The goal of the proposed collaborative research is to understand how distinct social behaviors are encoded in the genome and wired in the brain. Using a novel natural vertebrate system we will trace key behaviors to their genomic roots. Specifically, we will study innate 'bower' (reproductive nest) construction by male cichlid fishes from Lake Malawi, East Africa. This behavior is an essential part of social and reproductive interactions between males and females. In Aim 1, we will quantify the elements of bower behavior, comparing pit-digging vs. castle-building cichlid species and their F1 hybrids. Aim 2 will map brain regions recruited during male bower behavior by visualizing the expression of immediate early genes that reflect neural activity. Transcriptional divergence in these focal brain regions will then be quantified by allele-specific expression analysis in F1 hybrid males, who construct bowers in a phased sequence of species-specific behaviors. Aim 3 will identify regions of the genome responsible for differences in bower behavior by re-sequencing the genomes of bulked F2 individuals who either dig pits or build castles. Our integrative approach is designed to solve problems that are currently impossible, or at best difficult to study in standard inbred model organisms. Our strategy to identify the genetic control of neural circuits regulating essential behaviors in outbred vertebrates will provide a natural model of human behavioral diversity.